Admissions Is a Hard Problem

There was an interesting article on Inside Higher Ed yesterday about the idea of "Affirmative Action for Men." The piece was a response to an op-ed by Jennifer Delahunty Britz, an admissions officer at Kenyon College, where she talked about gender preferences in admissions, using the classic op-ed device of talking about a particular student she had rejected:

Had she been a male applicant, there would have been little, if any, hesitation to admit. The reality is that because young men are rarer, they're more valued applicants. Today, two-thirds of colleges and universities report that they get more female than male applicants, and more than 56 percent of undergraduates nationwide are women. Demographers predict that by 2009, only 42 percent of all baccalaureate degrees awarded in the United States will be given to men. [...]

The elephant that looms large in the middle of the room is the importance of gender balance. Should it trump the qualifications of talented young female applicants? At those colleges that have reached what the experts call a "tipping point," where 60 percent or more of their enrolled students are female, you'll hear a hint of desperation in the voices of admissions officers.

Beyond the availability of dance partners for the winter formal, gender balance matters in ways both large and small on a residential college campus. Once you become decidedly female in enrollment, fewer males and, as it turns out, fewer females find your campus attractive.

The Inside Higher Ed piece is mostly a summary of the predictable howls of outrage over this policy, from predictable quarters. I thought it was an interesting piece, and struggled a bit with how to write something sensible about it without sounding like a complete troglodyte. As always with controversial matters in academia, the correct answer is "wait until Timothy Burke writes something," but I'll put a few thoughts of my own below the fold.

The big question raised by this piece, and more importantly, by the reaction to it, is "How can the same people who support (or supported) admissions preferences for women and minorities get all torqued off about this policy without their heads exploding from the logical contradiction?" (OK, that's a slightly combative phrasing, but that's the idea...)

In the end, it comes down to what you think the proper purpose of affirmative action programs is. As Kate pointed out when we were talking about this last night, in the legal world, there have historically been two different justifications for affirmative action: remediation, and diversity. The remediation theory is that admissions or hiring preferences are necessary to compensate for past discrimination in admissions or hiring. The diversity theory is a little more nebulous, but basically holds that having a diverse group of students or employees is beneficial for the community as a whole, and obtaining those benefits is worth giving some preference to groups that would otherwise be underrepresented.

As I see it, there's no problem with supporting affirmative action generally but opposing Kenyon's preference for a 50/50 gender split if you're using the remediation theory. That version of affirmative action, as I see it, would be consistent with a floor, but not a cap. If under-representation of some group in the past has been due to discrimination only, there's no reason why the balance couldn't tip in the other direction, once the discrimination is gone. Maybe women are just innately better at academic pursuits than men, and the only thing that's prevented them from being 60% of the college-bound population in the past is the oppressive hand of the patriarchy.

This is explicitly the position taken by Katha Pollit, quoted in the Inside Higher Ed piece:

Pollitt also said that these policies debase affirmative action, which she strongly supports. "Affirmative action is intended to remedy past discrimination. There is no past discrimination against white males," she said. She also sees these policies as defining the college experience as social, not educational. "Is this an intellectual endeavor or the prom committee?"

There are a bunch of problems with this view, including the thorny questions it raises about how long you need to keep providing preferences to historically underrepresented groups, but at least it's fairly consistent. You may or may not agree with the underlying theory, but there's no logical contradiction between these views.

The diversity argument, on the other hand, strikes me as harder to reconcile with opposition to Kenyon's policy. The argument from diversity is basically that interacting with a large number of people from different backgrounds is a Good Thing for college students, and thus, admissions offices should strive for a mix of races, classes, and ethnic backgrounds that they think is appropriate. If they feel it's important for students to meet and interact with green-skinned people from Mars, it's OK to give a green-skinned person from Mars preference over people from Earth, if that's what you need to do to have your student body be 2% green-skinned people from Mars.

This is a somewhat more popular argument these days, as it avoids some of the more difficult questions raised in the remediation theory (in favor of airy generalizations about the wonders of diversity, which may or may not be an improvement). It's also been somewhat more successful in recent court cases, as I understand it.

If that's the justification for admissions preferences generally, though, then it strikes me as awfully hard to oppose Kenyon's policy. After all, that's what they're doing-- they've decided that it's important for the institution to have a class that's closer to 50/50 than 60/40, gender-wise, and they're taking steps to insure that. The fact that this involves giving a small preference to male applicants is ironic, but ultimately, sort of a footnote to the general policy goal. Or, as Burke puts it:

If you don't like this approach when it comes to gender, then arguably you don't like it when it comes to race, ethnicity, geographical origin, and even accomplishment when accomplishment is not directly connected to probable academic success. If you think this is valid on everything but gender (as Pollitt seems to), I'd like to hear how you see the difference.

Of course, it's also important to avoid blowing this whole thing out of proportion. The Kenyon policy, whatever it is, is actually a pretty small effect. As a commenter at Inside Higher Ed points out, the applicant pool is 55% female, and the entering class is 54% female-- that's not that big a deal, assuming that male and female applicants are equally well prepared. And, as Burke notes, this is one non-academic preference among many:

Better to be from Alaska than New York City. Better to be a first-generation college student-or a legacy. Better to be a person of color, unless you're of East Asian descent. Better to have a highly anomalous talent or background than be a valedictorian and student body president.

The ultimate lesson of all this is, well, the post title: deciding who to admit to college is a Hard Problem. There are lots of well-qualified applicants, and a limited number of spots, and at some point, hair-thin distinctions are going to have to be made. No matter what you decide, somebody is going to be unhappy.

Do I have any answers for this? No, and if I did, I certainly wouldn't waste them on a blog post-- I'd put them into practice, and go on to claim a well-deserved Nobel Peace Prize. I'm just thinking out loud. Applying this to the neverending "Women in Science" debate is left as an exercise for the reader.